I am a first year CSE grad student. I have been working with a professor for more than half a year now. The topic looks like a deadend and I do not know how to tell him the same. Ironically, he is a nice guy and has supported me all the way through.

My prof meets with me twice a week and brainstorms with me about possible avenues of this research. The sad part, I think, my prof is not sure if a particular line of research of work may be successful or not. I am not looking for a foolproof way of doing research. At least, 5 out of 10 attempts should be successful, I guess. Correct me, if I am wrong. Currently, I just feel like a mule trying to implement his ideas. There seems to be no theoretical justification for why a particular avenue could lead to good results. I could as well implement my own ideas and probably could have a better success rate.

This is not how I expected my project to turn out. Around 90% of the time, I just keep working on Code (writing code and fixing bugs), without any learning I can take away for future. Only 10% of the time, I just read some conference papers related to this topic. It is just too much of study in a small highly specialized area.

I love to understand new concepts and teach them to others, hence the reason I started out for Ph.D. (so that I can settle down as a Prof). Now I am not sure if I am preparing myself for a career in academia as most of the time I just keep working on Code - There is nothing new to learn so that I can teach to my future students.

I expected research to be more theoretically enriching but it is turning out to be the exact opposite.

These long hours have already taken a toll on me. I am missing out on the 'enjoy your research part'. When I started out, I had this zeal for research and enjoyed the work. It has been one year. Now, I just want to publish whatever results I have already with us, in any 2-tier journal/conference, without further go down this route.

But, my professor expects me to keep trying.

Sometimes, I feel I am better off with any supervisor. At least, I can study stuff which I like to.

  • 7
    Why do you assume a success rate of 5 / 10, some do hundreds of tests to find the one that works...
    – Solar Mike
    Jul 26, 2018 at 14:55
  • 2
    it is not the journey traveled, but the destination - find the correct answer is the pot of gold...
    – Solar Mike
    Jul 26, 2018 at 15:07
  • 1
    Sounds like you need to learn how to fail and how to take away lessons from that failure. Have you brought your concerns to your professor? Research isn't all great success and code well most of the time it's determining why the code fairies won't compile your instructions. Jul 26, 2018 at 15:30
  • 5
    "At least, 5 out of 10 attempts should be successful, I guess." That is stunningly unrealistic. Research involves discovering things that no human has done before. My lab and I recently brainstormed ~60 projects. I'd guess that many will end up dead ends; many will lead to new questions, even more interesting than the originals; a few will be straightforward & successful in the short term. But we're excited about all of the projects because, as noted in a good comment above, we are excited about the journey, not the destination. Research is very different than what you seem to think it is! Jul 26, 2018 at 15:36
  • 3
    my prof is not sure if a particular line of research of work may be successful or not — This is a good sign. If he were sure that the line of work would be successful, it wouldn't be research!
    – JeffE
    Jul 26, 2018 at 19:13

2 Answers 2


You are in a bad spot and it is difficult to diagnose the disease, much less the cure. It may be that your advisor is inexperienced - some things you say indicate that. But as commenters have implied, some problems worth solving are just hard - damn hard. I once worked on such a problem and had to abandon it, but the professor agreed and we found a better one.

However, it may be that your "best" solution is to find a different advisor. That can be difficult (or impossible) as there is department politics to consider. Some large departments even have factions. The radical solution is to change universities and thus find a different professor with different ideas.

But a student working on an extremely hard problem is itself a problem since it takes time and the student would like to actually graduate before the Sun goes nova. I also worked on problems that were too easy (mathematics) and hence had no real merit. The trick is to find the sweet spot of a sufficiently hard but do-able problem. Not all professors have the ability to to that regularly.

But you need to judge. If the problem is hard, but its solution would be a significant advance and you have the time and resources to pursue it, it can result in a big win. But the win isn't guaranteed.

However. It also sounds like your current advisor is truly invested in you. Two meetings a week shows commitment. You won't always find that in an advisor.

You might try two things. First is to have a conversation about the likelihood of actual success in a reasonable time frame. The other part of the conversation is to try to set a limit on attempts and explore alternative problems that don't completely ignore what you have done so far. If your advisor is, in fact, invested in you and not just in the problem itself, that conversation might be possible.

As a first year doctoral student it isn't yet time to panic, in fact. Significant things take time and focus.


You sound like a perfectly normal PhD student.

Your supervisor sounds okay. At least he shows up for meetings!

Look out for your own interests, but don't worry so much.

You must log in to answer this question.